The biliary system is one of the most complicated systems of the human organism [1]. It entails an extrahepatic and an intrahepatic component, which seem to differentiate separately [1]. Both components eventually join in a unique excretory system of bile drainage. The extrahepatic biliary system results from a process intimately associated with pancreatic embryogenesis [2]. Conversely, the intrahepatic component derives from the periportal areas of the ductal plate, and the co-localization of polyductin with liver progenitor cell markers is critical ^[3][4][3,4]. The ductal plate is the prototype of embryogenesis ^[5][6][5,6]. If the joining of the intra- and extrahepatic components fails, such disconnection results in atresia of the biliary drainage system (BA) ^{[1][3][4][5][7]}[1,3,4,5,7]. Biliary atresia is a very complex disorder and, in the most recent years, the adjective fibro-obliterative has slowly replaced the previously accepted term, necro-inflammatory ^[8][9][10][8,9,10]. Bove et al. reported detailed histology in 172 centrally reviewed biliary remnants performed at Kasai hepatic portoenterostomy (KHPE). Active lesions were classified as either necro-inflammatory or active concentric fibroplasia with or without inflammation, whereas inactive lesions disclosed bland replacement by collagen and fibrous cords with little or no inflammation. ResWearchers need to acknowledge that heterogeneity is a common and substantial finding in BA. According to Bove et al., it did not correlate with age at KHPE, congenital anomalies at laparotomy, and outcome. Prevalence of partially preserved epithelium in active fibroplastic lesions may indicate that epithelial damage (regression or injury) may not be a primary event. Alternatively, the re-epithelialization is already in progress at the time of KHPE. The unfathomable balance between active fibroplasia and re-epithelialization of retained, collapsed but not obliterated lumens is remarkably intriguing. At places, it may explain the heterogeneity of BA histopathology reports [11].

It has also been suggested that BA is the consequence of repeated bouts of the inflammation destined to clear the toxins derived by the malfunction of the bile drainage system ^[3][7][10][3,7,10]. The liver hilum’s surgical and pathological inspection shows, indeed, an obstructive cholangiopathy. There is an annihilation of the lumen and progressive iterative cycles of inflammation with an overwhelming NLRP-3 cytokine response ^{[10][12][13][14]}[10,12,13,14]. BA may include situs viscerum inversus and is labeled BA splenic malformation syndrome (BASM). Its clinical presentation typically occurs in the 2nd week after birth ^[15][16][15,16]. A variant of BA is the cystic BA (CBA), which has shown a better outcome than the classic embryonic and fetal types and does not seem to require liver transplantation.

The search for the perfect animal model goes hand in hand with the quest to identify the etiology of BA. Substantially, there have been two lines of thought: one school has deposited trust in a viral etiology, whereas another school sustains that genetic mechanisms are probably the underlying disorders with or without the influence of toxins. The existence of viral-induced BA in mice has been often extrapolated to suggest that viral-mediated injury must play a role in human BA, but extensive human data have failed to support such a hypothesis. In the viral etiology, there is the concept that an immune or autoimmune response to the perinatal infection of a virus may lead to necrosis of the biliary epithelium. Cytomegalovirus (CMV) ^{[17][18][19][20]}[17,18,19,20], reovirus ^[21][22][23][21,22,23], rotavirus ^[21][24][21,24], Epstein-Barr virus ^[21][25][26][21,25,26], or human papillomavirus (HPV) ^[20][21][20,21] can induce a subsequent inflammation and/or accompany an evidence of inflammation. Such infection is followed by a progressive fibro-obliteration of the lumen of the interlobular bile ducts. Some culpable viruses have been published in case reports only, but extensive human data have failed to validate such a hypothesis, which was never substantially confirmed as causative in case-control studies or cohort studies. The ductal plate malformation (DPM), which is the archetypical defect of the primitive biliary system, has been identified in some cases with BA associated with splenic anomalies ^{[4][7][27][28][29][30][31][32][33][34][35]}[4,7,27,28,29,30,31,32,33,34,35]. The complexity of the degree of inflammation and state of the development of the biliary system in BA, combined with the urgent need for new therapeutical approaches, requires well-defined animal models that enable the translation of research findings to clinical use.

2. Experimental Animal Models Conundrum

Some rodent models are well suited for studying mechanisms of various aspects of inflammation. Still, they have shown limited utility for translating genetic research data to human terms, thereby impeding therapeutic developments. There is a need to emphasize the need for standardization of preclinical animal models by showing the impact of duration of jaundice, sex, and methodologies. Recognizing and realizing the limitations and advantages of preclinical models will assist a more effective translation of experimental results to enhanced approaches to therapy for human BA. This neonatal cholangiopathy proceeds through several overlapping stages that unequivocally necessitate precise spatiotemporal control. It also needs contributions from multiple cell types that orchestrate the myriad of cellular processes that eventually result in BA. This extraordinary process is evolutionarily conserved and can be studied in numerous in vivo models, ranging from zebrafish to primates. Over the past several decades, the use of BA models has contributed to breakthroughs, advancing the knowledge and interpretation of the molecular and cellular events that contribute to this process. Despite advances in the biology of understanding BA, understanding the pathophysiology of acute and chronic inflammation in the developing liver is limited, resulting in a lack of successful therapies for the affected infants. Animal models that have been designed to investigate biology and preclinical mechanisms resulted in variations that have not been properly standardized and validated, creating a challenging situation. Several variables influence this already complex biologic process. They include age, sex, biliary development, metabolic pathways, and microbiome diversity. Consequently, reswearchers encounter a conundrum. On the one hand, researcherswe face the situation where the field is required to use animal models to perform preclinical investigations as a necessity for therapeutic development. On the other hand, researcherswe are often confronted with the limited predictive value of data obtained in preclinical examination for clinical implementation. It is critical to understand both the limitations and advantages of various models. Accurate experimental animal models may allow the understanding of pathophysiological mechanisms involved. Different animal species like the rat, lamprey, and rhesus monkey have been used, but murine models have been used more often. The reproduction of BA requires various strategies, including surgical methods (e.g., common bile duct (CBD) ligation), viral infection, grafting, and drug administration. ResearchersWe will describe the available animal models, their development, applications, and practical considerations for scientific usage.

3. Non-Human Primates and Non-Primates Occurrence of BA

Spontaneously occurring BA-like findings have been reported in lambs, dogs, calves, and foals but have not been adopted for studies. The more investigated spontaneous BA models in animals are zebrafish, lamprey, and Rhesus monkey and animal models for BA can be subdivided in viral, surgical, toxin-based, and genetic (Figure 1).

4. Zebrafish

Although distant from the human condition, the tropical freshwater zebrafish (Danio rerio) may be used as a model for studying biliary defects due to the remarkable conservation of development genes. Since the 1960s, the zebrafish model has become increasingly important to scientific research. This tropical fish native to Southeast Asia has many characteristics that make it a valuable model for studying human genetics and disease. The complete genomic sequence of the zebrafish was published in 2013, and its genome reaches 1,505,581,940 base pairs in length (26,247 protein-coding genes) ^[36][37][38][36,37,38]. The notch pathway, which is essential in mammalian biliary development, also regulates the specification of liver progenitor cells towards a biliary cell fate in zebrafish [39]. In the zebrafish model of BA, the inhibition of DNA methylation leads to biliary defects and activation of interferon gamma (IFN-γ) genes. Activating the IFN-γ pathway is sufficient to cause developmental defects. Subsequently, there is a decrease in cholangiocyte proliferation and expression of vhnf1 (hnf1b, tcf2) ^[40][41][42][40,41,42]. In addition, BA patients show, in the cholangiocytes, a reduction of the levels of apical glypican-1 (GPC1) ^{[43][44][45][46][47][48][49]}[43,44,45,46,47,48,49], a heparan sulfate proteoglycan involved in regulating hedgehog signaling. Knockdown of gpc1 in zebrafish leads to biliary defects, which can be partially reversed with cyclopamine, a hedgehog antagonist. Most recently, HNF6, ARF6, Sox17, ILF2, DNMT1 genes have been confirmed to have effects on the development of bile ducts in zebrafish ^{[46][50][51][52][53][54][55][56][57][58]}[46,50,51,52,53,54,55,56,57,58]. In particular, the deletion of ilf2 determined a reduction of bile flow and defect in the biliary anatomy, which has been interpreted as a phenocopy of BA in humans. It is imperative to investigate the embryonic biliary development in Zebrafish in more detail and compare BA development as induced by virus (RRV) or toxins (biliatresone) in heterozygous/homozygous ilf2 knockout rodents to further illuminate the roles of ILF2 in the human embryonic biliary development and BA [50]. In Table 1 are detailed the most paramount animal models of biliary atresia.

5. Lamprey

The Sea Lamprey (Petromyzon marinus) is a primitive, eel-like fish native to the northern Atlantic Ocean and freshwater of the Great Lakes ^{[59][60][61][62][63]}[59,60,61,62,63]. The sea lamprey metamorphosis is a genetically pre-programmed animal model for BA ^{[64][65][66][67][68][69][70][71][72][73]}[64,65,66,67,68,69,70,71,72,73]. The larva has a fully formed biliary system with the gallbladder, lost during the transition to the adult stage. The epithelium of the extrahepatic ducts in lamprey larvae transforms and expands into a caudal portion of the pancreas of the adult. Regression of the biliary structures is accompanied by periductular fibrosis. Although Youson initially noted these morphological similarities in 1978 ^[66][74][66,74], the lamprey was not adapted as an experimental BA model until recently in this century ^{[59][64][65][67][68][69][70][71][72][73][75]}[59,64,65,67,68,69,70,71,72,73,75]. In human BA, the most severe injury seems to occur early in the extrahepatic tract, and the intrahepatic ducts show ductular proliferation. In the lamprey, the degenerative process is asynchronous. The more rapid decline occurs in small peripheral components, and post-metamorphosis lampreys usually grow to adult size without developing any sort of progressive disease, despite having no bile ducts. Yeh et al. reported that the sea lamprey adapts by de novo synthesis and secretion of bile salts in the intestine and reduction of bile salt synthesis in the liver [71]. According to Cai et al. [70], adult sea lamprey tolerates BA by transforming its bile salt composition from toxic C24 bile acids (petromyzonol sulfate) to less harmful C27 bile acids (3-keto-petromyzonol sulfate). It accelerates renal excretion through urine, supported by marked upregulation for organic anion and bile acid transporters. Further studies may be necessary to explore, validate, and leverage this BA model.

6. Rhesus Monkey

Rhesus monkey represents a vital compromise to have an animal model that is close to humans and more feasible for surgical manipulation [76]. The only pure extrahepatic biliary atresia known in a nonhuman primate is identified in the rhesus monkey (Macaca mulatta) neonates. In 1983, Rosenberg et al. showed clinical and pathologic similarities to BA in a six-week-old female monkey [77]. Jaundice and conjugated hyperbilirubinemia at the age of six days persisted throughout life with the portal proliferation of biliary structures. The autopsy revealed biliary cirrhosis with ductular proliferation at ten months. High reovirus 3 titers in rhesus neonates’ serum were also found in human infants with BA ^[23][78][23,78]. In 1974, Landing first hypothesized that hepatotropic viruses play an etiologic role in BA, but numerous studies have substantially annihilated the viral hypothesis. It is important to keep in mind that no detailed studies have been performed on the rhesus neonate animal model due to the costs and difficulty in obtaining Ethics Committee approval.